·
5 commits
to main
since this release
Immutable
release. Only release title and notes can be modified.
Users can now use ocr as well, built using PaddleOCR, and adding features like annotation and translation and adding boundaries for detected text and translating them into english and displaying them in a new image.
Usage:
vllama ocr -p "image_path" -a 'for annotation and translation' -o "output_directory" -l "language of text in the image"